Bayesian Adaptive Tactical Decision Support System for Manned-Unmanned Teaming Operations

This paper presents the Bayesian Adaptive Tactical Decision Support System designed to support operators in Manned-Unmanned Teaming (MUM-T) operations. The system integrates a situation assessment model that uses a Bayesian Neural Network (BNN) for real-time situation prediction and awareness in a probabilistic manner, and a decision-making model that generates adaptive Courses of Action (COAs) based on probabilistic evaluations from the BNN. By leveraging Bayesian methods, the system can handle uncertainty and dynamic environments, providing operators with reliable decision support and actionable recommendations. The proposed approach enhances situational awareness and decision-making in complex, high-risk military operations. Simulation results demonstrate the system's effectiveness in improving mission planning and tactical responses. Future work will focus on scalability, real-world data integration, and further validation in diverse operational contexts.